View metadata, citation and similar papers at core.ac.uk brought to you by CORE + MODEL ARTICLE IN PRESS EJSOBI2173_proof 7 March 2007 1/8 provided by University of Debrecen Electronic Archive 1 53 2 54 3 55 e 4 European Journal of Soil Biology xx (2007) 1 8 56 http://www.elsevier.com/locate/ejsobi 5 57 6 Original article 58 7 59 8 Changes of isopod assemblages along an urbane 60 9 e 61 10 suburban rural gradient in Hungary 62 11 63 12 Elisabeth Hornung a,*,Be´la To´thme´re´sz b, Tibor Magura c, Ferenc Vilisics a 64 13 65 a 14 Department of Ecology, SzIU, FVS, Institute for Zoology, H-1400 Budapest, PO Box 2, Hungary 66 b Ecological Institute, University of Debrecen, Hungary 15 c Hortoba´gy National Park Directorate, Debrecen, Hungary 67 16 68 17 Received 13 December 2005; acceptedPROOF 17 January 2007 69 18 70 19 71 20 72 21 Abstract 73 22 74 23 Responses of isopod assemblages to urbanisation were studied along an urbanesuburbanerural gradient representing a decrease 75 24 in the intensity of human disturbance. Pitfall trapping collected six species (Armadillidium vulgare, Porcellio scaber, Porcellium 76 25 collicola, Trachelipus ratzeburgii, Cylisticus convexus, and Trachelipus rathkii). A. vulgare occurred abundantly in all sites reflect- 77 ing the broad tolerance and invasive nature of this species. Indicator species analysis demonstrated that P. scaber and T. rathkii were 26 78 significant quantitative character species for the urban site, while T. ratzeburgii was characteristic for the natural habitats (suburban 27 and rural sites). CANOCO revealed that ground and air temperature show positive correlation with the distribution of P. scaber and 79 28 T. rathkii, and negative correlation with T. ratzeburgii. Nested ANOVA on trap level showed that there were no significant differ- 80 29 ences between the number of isopod species and individuals, and the diversity of isopod assemblages in the three studied areas. 81 30 Significant differences were observed at site level. The results did not support the hypothesis that diversity should decrease in re- 82 31 sponse to habitat disturbance. They also contradicted the intermediate disturbance hypothesis; species richness was not the highest 83 32 in the moderately disturbed suburban area. Multivariate methods detected that the isopod assemblages of the rural and suburban 84 33 areas were relatively similar, while that of the urban area was relatively separated. 85 34 Ó 2007 Elsevier Masson SAS. All rights reserved. 86 35 87 36 Keywords: Oniscidea; Globenet; Increased disturbance hypothesis; Intermediate disturbance hypothesis; Species richness; Diversity; Urbanisation 88 37 89 38 90 39 1. Introduction to suburban and rural forests, including higher air pollu- 91 40 tion and disturbance intensity, the heat island phenom- 92 41 The effects of urbanisation can be explored through enon and the presence or greater abundance of exotic 93 42 investigations of biotic and abiotic changes along species [34,41]. The floristic richness of many urban 94 43 urban-to-rural gradients [27,29]. Such gradients, from habitats frequently exceeds that of less urbanised areas 95 44 densely built inner citiesUNCORRECTED to increasingly rural surround- [45], reflecting the diverse, and mosaic nature of urban 96 45 ings, reflect diminishing intensities of human influence. habitats and the presence of introduced plants. The 97 46 Urban forests are exposed to unique features compared cityeforest ecotone also plays an important role in 98 47 maintaining this diversity [5]. 99 48 100 * Corresponding author. Tel.: þ36 1 478 42 33; fax: þ36 1 478 42 32. Recently, a multi-national research framework has 49 E-mail address: [email protected] (E. Hornung). been initiated to assess and compare the influence of 101 50 102 51 1164-5563/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved. 103 52 doi:10.1016/j.ejsobi.2007.01.001 104 Please cite this article in press as: E. Hornung et al., Changes of isopod assemblages along an urbanesuburbanerural gradient in Hungary, Eur. J. Soil Biol. (2007), doi:10.1016/j.ejsobi.2007.01.001 ARTICLE IN PRESS EJSOBI2173_proof 7 March 2007 2/8 2 E. Hornung et al. / European Journal of Soil Biology xx (2007) 1e8 105 urbanisation using invertebrates and standardised field (47 350 N21 370 E). The criteria for distinguishing 157 106 methods (Globenet) [30]. Carabids were selected as urban, suburban and rural area were the ratio of the 158 107 the focal taxon, but other taxa (ants, and spiders) have built-up area to the natural habitats. Disturbance was 159 108 also been studied recently within the Globenet frame- estimated based on a 1  1 km unit around the centre 160 109 work [1]. According to the Globenet protocol, we in- of the investigated area. In the urban area the built-up 161 110 volved in our research three kinds of forested habitats area exceeded 60%, in the suburban area it was approx- 162 111 (urban park, suburban forested area, and rural forest), imately 30%, while in the rural area the built-up area 163 112 representing different levels of human disturbance was 0%. The area of the built-up environment and the 164 113 [31]. The target taxon was the terrestrial, surface active natural habitats was measured by the ArcView GIS 165 114 woodlouse assemblage. Isopods are regarded as a useful program using an aerial photograph. 166 115 and reliable monitoring group. They are widespread, 167 116 easily identified and in most cases dominant compo- 2.2. Sampling design 168 117 nents of the macrodecomposer guild in temperate 169 118 regions [32]. Isopods belong to the saprophagous Four sites, at least 50 m apart, were selected within 170 119 components of the soil macrofauna. These organisms each sampling area (urban, suburban, and rural). Iso- 171 120 process the majority of dead organic material and so pods were collected at each site using pitfall traps, ran- 172 121 influence the rate of nutrient release, an important eco- domlyPROOF placing 10 traps at least 10 m apart from each 173 122 system function. The urban fauna can be very diverse, other at each site. This resulted in a total of 120 traps 174 123 often due to introduced, exotic species [18,20]. In this scattered along the urbanerural gradient (3 area  4 175 124 paper, we tested the following predictions for isopods sites  10 traps). The pitfall traps were unbaited, con- 176 125 in urban environments: (1) diversity should be highest sisting of plastic cups (diameter 65 mm, volume 177 126 in the suburban area (IDH e intermediate disturbance 250 ml) containing 75% ethylene glycol as a killing- 178 127 hypothesis); and (2) diversity should decrease from preserving solution. The traps were covered with bark 179 128 a high value in the rural area to a low one in the urban pieces to protect them from litter and rain. Trapped iso- 180 129 area (increasing disturbance hypothesis). We also inves- pods were collected from the end of March to the end of 181 130 tigated the changes in the isopod assemblages along the November 2001. For the purpose of analysis, we pooled 182 131 urbanisation gradient, identified the characteristic and/ samples from the whole year. 183 132 or key species across this gradient, and correlated cer- Ground temperature at 2 cm depth, and air tempera- 184 133 tain environmental variables with the observed pattern ture and relative humidity at the soil surface were mea- 185 134 of isopod abundance and species richness. sured adjacent to each trap monthly on the morning of 186 135 a typical sunny day. The statistical analyses were based 187 136 2. Material and methods on averages. We also estimated the percentage cover of 188 137 leaf litter, decaying wood material, herbs, shrubs and 189 138 2.1. Study area the canopy within a radius of 1 m around each trap 190 139 (see Table 1). 191 140 The study areas were situated in and around the city 192 141 of Debrecen (Eastern Hungary), the second largest city 193 142 of the country. Three forested sampling areas e each 194 143 covering an area of at least 6 ha e were selected along 195 Table 1 144 an urbanisation gradient [16]; this represented urban, Average values (ÆSE) of the studied environmental factors in the 196 145 suburban and rural areas, according to the Globenet pro- study areas 197 146 tocol [30]. All sampling sites were in forest stands dom- Urban Suburban Rural 198 147 inated by English oak (Quercus robur); distance among 199 Ground temperature 24.9 Æ 0.249a 22.3 Æ 0.100b 21.6 Æ 0.249c 148 the studied areas wasUNCORRECTED at least 1 km [24]. In the urban Air temperature 31.2 Æ 0.146a 27.3 Æ 0.076b 27.8 Æ 0.220c 200 149 park area, there were several asphalt-covered paths Relative humidity 60.4 Æ 0.744a 76.6 Æ 0.495b 58.9 Æ 0.503c 201 150 and the shrub layer was strongly thinned, while in the Cover of leaf litter 21.1 Æ 4.152a 57.1 Æ 4.181b 21.1 Æ 3.207a 202 a a b 151 suburban area the fallen trees were removed, while in Cover of decaying 3.8 Æ 0.495 4.2 Æ 0.557 11.0 Æ 1.442 203 wood material 152 the rural area forest management was only occasional a b c 204 e Cover of herbs 46.5 Æ 5.243 29.1 Æ 4.108 68.6 Æ 3.348 153 at a low-intensity level.